registry / @markjaquith/agency / 2.73.2

@markjaquith/agency@2.73.2

Manage agentic work across repositories with durable workbases

AI Security Review

scanned 14d ago · by lpm-firewall-ai

LPM treats this as warn-only first-party agent extension lifecycle risk. npm postinstall copies a packaged Pi extension into the user's Pi extension directory. At Pi agent startup, that extension can add Agency workbase instructions to the agent prompt.

Static reason

One or more suspicious static signals were detected.; previous stored version diff introduced dangerous source

Trigger

npm install, then a Pi agent session in an Agency workbase

Impact

Changes Pi agent behavior for Agency workbases; no confirmed exfiltration or remote execution.

Mechanism

install-time Pi extension deployment and prompt-context hook

Rationale

The package performs an unprompted first-party Pi extension installation and prompt augmentation, which warrants a warning under the lifecycle policy. Its scoped implementation and lack of concrete malicious behavior do not support a block.

Evidence

package.jsonscripts/install-pi-extension.tspi-extensions/agency.tssrc/utils/interactive-loader.tssrc/usage-log.ts$HOME/.pi/agent/extensions/agency.ts

Decision evidence

public snapshot

AI called this Suspicious at 91.0% confidence as Dangerous Capability with low false-positive risk.

@markjaquith/agency@2.73.2

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Source & flagged code